AHF Balance Cluster Report ¶

Recommended CPU_COUNT values ¶

Notes:

• vCPUs: The sum of the CPU_COUNT values for all instances on each machine as compared to the number of vCPUs available on that machine. Variation of these numbers across the cluster is indicated with square brackets [].
• Exposed(hrs): The number of hours during which any database instance in the cluster was exposed as compared to the total number of hours for which metrics were collected.
• Impacted(hrs): The number of hours during which any database instance in the cluster was impacted as compared to the total number of hours for which metrics were collected.
• Recommended values show what would have happened if the recommended CPU_COUNT values had been in place.

Best practice is for total configured CPUs to be less than or equal to physical CPUs.

The graph below shows when Exposure and Impact occurred (the Current Scenario), and how the use of the recommended CPU_COUNT values would have changed that picture (the Recommended Scenario). The horizontal axis is time, from left to right. The colors in each individual stripe indicate the status at each point in time for a particular Scenario. Hovering over the graph with the mouse shows the details for a single data point.

Here is the distribution of Exposure and Impact across machines:

Notes:

• vCPUs: The sum of the CPU_COUNT values for all instances on each machine as compared to the number of vCPUs available on that machine.
• Exposed(hrs): The number of hours during which any database instance running on the machine was exposed as compared to the total number of hours for which metrics were collected.
• Impacted(hrs): The number of hours during which any database instance running on the machine was impacted as compared to the total number of hours for which metrics were collected.
• Recommended values show what would have happened if the recommended CPU_COUNT values had been in place.

The graph below has a pair of horizontal stripes for each machine.

Database Details ¶

Notes:

• vCPUs: The CPU_COUNT values for all instances of each database as compared to the number of vCPUs available on that machine where it is running. Variation of these numbers across the instances of the database is indicated with square brackets [].
• Exposed(hrs): The number of hours during which any instance of the database was exposed as compared to the total number of hours for which metrics were collected.
• Impacted(hrs): The number of hours during which any instance of the database was impacted as compared to the total number of hours for which metrics were collected.
• Recommended values show what would have happened if the recommended CPU_COUNT values had been in place.

The graph below has a pair of horizontal stripes for each database.

Data Warnings ¶

Missing Data Warnings ¶

Please note:

Metrics may be missing for a variety of reasons. The database or host may be down or in blackout, for example, or the EM agent may be unreachable. If metric data is available for most of the time interval, the values given for Exposure and Impact will be reasonably accurate. However, if metric data is missing for much of the time interval, the actual Exposure and Impact values may be higher than those reported here, depending on the reason that the data is missing. Missing database instance metrics affect the Exposure values, while missing host metrics affect the Impact values.

The tables below list databases and hosts where metrics were missing for more than 10% of the samples. If you need more accurate Exposure and Impact values for the database instances listed below, or for database instances running on any of the hosts listed below, the metric collection problem needs to be found and fixed. Please use this guide as a reference. Once metrics are being collected successfully, and enough time has elapsed to collect new metrics, this report can be re-run.